Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This should fix generation + eval tasks for pipeline parallel sizes > 1 in conjunction with model parallel > 1
I haven't yet compared 20B outputs when generating to the main branch, since I don't have access to any A100 pods rn. Would appreciate if someone could do that, but eval results remain the same.
main:
PR:
Changes:
ParallelTransformerLayer
class.get_key_value
touse_cache
, since I found that naming very confusing